Agents heat 2 · Cool★ 1heat incomplete (no X)

jev-claud

Python library + Claude plugin: keep free-form text for people, route Choice/Score/Noul decisions to TypeSafe Jev (PyPI, CI, CLI).

What it is

Typed branching for Claude sessions without turning Jev into a chat model.

What it does: Gives code decide/when/rank helpers so uncertain questions hit System One once.

Why it belongs in the showcase: Shipped package with CI, privacy notes and Claude plugin wiring.

Caveats: PT docs; API key; Claude tokens remain.

Evidence

Public MIT repo (1★ at intake, size ~1195) with PyPI package jev-claud, GitHub Actions CI, install.sh/ps1, plugin/, examples/, PRIVACY.md. README (Portuguese/EN API surface) shows Case/choice/noul/score helpers, TYPESAFE_API_KEY, model aliases latest/preview, optional image/box helpers, and CLI jev-claud ask. Pushed through 2026-10-16.

Caveats

  • Primary docs language is Portuguese; API names stay English
  • Still needs a TypeSafe key for hosted Jev calls
  • Claude plan tokens still apply to Claude's own prose

Unofficial case page. Stars / list position observed for heat; do not treat heat as a quality SLA. Review the upstream README before you depend on it.